Construction and Building Materials 48 (2013) 499–507

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

Combined effects of recycled aggregate and fly ash towards concretesustainability

0950-0618/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.conbuildmat.2013.07.014

⇑ Corresponding author. Address: School of Urban and Environmental Engineer-ing, Ulsan National Institute of Science and Technology (UNIST), 100 Banyeon-ri,Eonyang-eup, Ulju-gun, Ulsan 689-798, Republic of Korea. Tel.: +82 52 217 2814.

E-mail address: msshin@unist.ac.kr (M. Shin).

Kyuhun Kim a, Myoungsu Shin a,⇑, Soowon Cha b

a School of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, Republic of Koreab Department of Civil and Environmental Engineering, University of Ulsan, Ulsan, Republic of Korea

h i g h l i g h t s

� The higher ratio of recycled aggregate resulted in the better flowability of concrete.� Recycled aggregate concrete with fly ash presented slightly lower yield stress, but much lower plastic viscosity.� Fly ash caused only a minor reduction in the strength of recycled aggregate concrete.� Recycled aggregate concrete with fly ash showed much higher chloride resistance than that without fly ash.

a r t i c l e i n f o

Article history:Received 7 March 2013Received in revised form 6 June 2013Accepted 15 July 2013Available online 9 August 2013

Keywords:Recycled aggregateFly ashRheologyStrengthChloride diffusion coefficientICAR Rheometer

a b s t r a c t

The recycling of demolished concrete has been emerging as a sustainable solution to warrant the reduc-tion of construction wastes, as well as to prevent the depletion of natural resources from growing con-struction demand. Nevertheless, some key factors that would affect the properties of recycledaggregate concrete have not been thoroughly investigated, such as the proportion of recycled aggregates,the moisture state of recycled aggregates, and the design compressive strength of concrete. In particular,little research was done on the combined effects of recycled aggregates and fly ash, popularly used as apartial substitution of cement. Given the concerns, this study investigates the effects of such factors onthe mechanical and durability properties of recycled aggregate concrete. Eleven cases of concrete mix-tures were tested for the rheological properties of fresh concrete, compressive strength, tensile strength,and chloride diffusion coefficient. In general, the higher ratio of recycled aggregates resulted in the betterflowability of concrete. Also, the use of fly ash improved the flowability of recycled aggregate concrete.The strength test results showed that the higher ratio of recycled aggregates generally caused the lowercompressive and tensile strengths of concrete. However, the cases with 30% recycled aggregates showedonly slight compressive strength reductions. Similarly, the use of fly ash caused only small reductions inthe compressive strength of recycled aggregate concrete. In contrast, the negative effects of recycledaggregates and fly ash were greater in the tensile strength than in the compressive strength. Lastly,the cases containing fly ash exhibited much higher resistance to chloride penetration, even in the caseswith recycled aggregates.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction and background

Climate change, which is often signified by global warming, isone of the most critical global issues that have potential to jeopar-dize the sustainability of human society. Among many causes, theconstruction industry is responsible for a major portion of green-house gas emission. For example, the production process of cementitself yields approximately 7% of the total CO2 emission worldwide[1].

Also, construction wastes take nearly 50% of the total solidwaste in the US [2], most of which are deposited into landfills thatmay cause serious environmental pollution problems in both localand global scales. On the other hand, the consumption of aggre-gates as a construction material has continually increased withuniversal industrialization and urbanization (e.g., expansion andupgrade of civil infrastructure). Moreover, enhanced legal con-straints for the conservation of natural resources have resulted inan imbalance between the demand and supply of aggregates inmany countries [3].

To meet the global consensus of sustainable development, tech-nical strategies for ensuring the sustainability of concrete con-struction have been discussed among the professionals [4–6],

Nomenclature

f 0c design compressive strength of concrete (MPa)s shear stress (Pa)s0 yield stress (Pa)l plastic viscosity (Pa s)_c shear strain rate (1/s)D non-steady-state migration coefficient (m2/s)z absolute value of ion valence (=1 for chloride)F Faraday constant (=9.648 � 104 J/V mol)R gas constant (=8.314 J/K mol)

T average of initial and final temperatures in the anolytesolution (K)

xd average chloride penetration depth (m)t test duration (s)U absolute value of the applied voltage (V)L thickness of specimen (m)erf-1 inverse of error functioncd chloride concentration at which the color changes (N)c0 chloride concentration in the catholyte solution (N)

500 K. Kim et al. / Construction and Building Materials 48 (2013) 499–507

which suggest saving materials in design, maximizing concretedurability, use of waste or supplementary cementitious materials(e.g., fly ash, and slag), and recycling of concrete. In this study,the last two approaches are investigated: the main objective is toinvestigate the combined effects of recycled aggregates and flyash on the mechanical and durability properties of concrete.

The recycling of demolished concrete, which comprises roughly50% of the total construction waste in South Korea, has beenemerging as a sustainable solution to warrant the reduction of con-struction wastes, as well as to prevent the depletion of natural re-sources (i.e., stones) from growing construction demand [7,8].Demolished concrete can be processed into recycled aggregatesthat may be used as a partial or complete replacement of naturalaggregates in concrete production. However, the application ofrecycled aggregates in concrete products is very limited, becauseof their inferior qualities to natural aggregates. As shown inFig. 1 [9], almost 95% of recycled aggregates are used for roadembankment, base, pavement, backfill, and such, while applica-tions to concrete products are less than approximately 5% in Korea.Korean Concrete Standard Specifications [10] state that, if theproperties and the particle size distribution are satisfied with sug-gested conditions in Table 3 and Fig. 2, the use of recycled aggre-gates is allowed within 30% of the total amount of aggregates forconcrete products that have the compressive strength of 21–27 MPa.

Recycled aggregate concrete generally shows a lower strengthand durability than natural aggregate concrete due to the higherporosity and lower density of recycled aggregates. The InterfacialTransition Zone (ITZ) existing between an aggregate and mortarin concrete is known as a critical factor affecting the overall qualityof concrete. Concrete made with recycled aggregates includes oldITZs (between the original virgin aggregates and residual mortar

Fig. 1. Usage of recycled aggregates in Korea [9].

attached to them) as well as new ITZs. This is presumed to bethe primary cause of its inferior properties [11].

Nevertheless, some key factors that would affect the qualities ofrecycled aggregate concrete have not been thoroughly investi-gated, such as the proportion of recycled aggregates, the moisturestate of recycled aggregates, and the design compressive strengthof concrete. In particular, little research was done on the combinedeffects of recycled aggregates and fly ash that is popularly used as apartial substitution of cement. Given the concerns, this studyinvestigates the effects of such factors on the mechanical and dura-bility properties of recycled aggregate concrete: the rheologicalproperties in the plastic state, compressive and tensile strengths,and chloride penetration resistance of concrete. The main objectiveof this study is to investigate the combined effects of recycledaggregates and fly ash.

Several studies found that the higher ratio of recycled aggre-gates had a tendency to increase the slump of fresh concrete[12–14]. Also, Siddique [15] tested with changing the substitutionratio of cement by fly ash up to 50%, and reported that the flowabil-ity of concrete enhanced with an increased amount of fly ash. Dur-ing a slump test, the flow of fresh concrete will stop when theshear stress generated by gravity in the concrete is smaller thanthe yield stress of the concrete. Based on this theory, there hasbeen much research to find out the correlation between slumpand yield stress. According to Wallevik [16], several equationswere proposed to elucidate the relationship of the two. However,the rheological properties (e.g., yield stress, and plastic viscosity)of recycled aggregate concrete have not been studied much untilnow, and may not be measured with the conventional slump tests.

Numerous studies investigated the effect of recycled aggregateson the strength of concrete [17–20], and the results generallyagreed that the strength decreased when the ratio of recycled

Fig. 2. Particle size distributions of coarse and fine aggregates used in the tests.

K. Kim et al. / Construction and Building Materials 48 (2013) 499–507 501

aggregates increased. According to Sim and Park [22], the strengthof concrete made with fly ash was smaller than that of concretewithout fly ash. Berndt [21] also examined the effect of a partialreplacement of cement with fly ash and/or blast furnace slag onthe properties of concrete. It was noted that natural aggregate con-crete with 50% of cement substituted by blast furnace slagachieved the best performance, while concrete containing fly ashshowed a lower strength and durability. Although separate effectsof recycled aggregates or fly ash were extensively investigated, lit-tle research has assessed their combined effects on the strength ofconcrete to date.

Berndt [21] examined the effect of recycled aggregates on thechloride diffusion coefficient of concrete. He reported that recycledaggregate concrete was more vulnerable to chloride attack thannatural aggregate concrete. Kong et al. [11] also found that theuse of recycled aggregates caused a lower resistance to chloridepenetration. In addition, Berndt [21] showed that the chlorideresistance of concrete was weakened with the use of fly ash. How-ever, the tests were conducted at the age of 28 days only, and noconsideration was given to the delayed development of concretewith fly ash.

2. Experimental procedures

2.1. Test variables

In this study, various properties of recycled aggregate concrete are investigatedsuch as the rheological properties of fresh concrete, compressive strength, tensilestrength, and chloride ion diffusion coefficient. The test variables are the ratio ofrecycled aggregates in the total amount of coarse aggregates, the presence of flyash as a partial substitution (30%) of cement, the design compressive strength ofconcrete, and the moisture state of recycled coarse aggregates. A total of elevencases of concrete mixtures are tested, as summarized in Table 1.

As for the designation of the mixture cases in Table 1, the first letters ‘‘N’’ and ‘‘H’’indicate that the design compressive strength of concrete is equal to 24 and 40 MParespectively, which represent normal and high strength concretes in turn. The addi-tional letter ‘‘F’’ means that 30% of the total cement is substituted by fly ash; in thecases without ‘‘F’’, fly ash is not included. Also, ‘‘OD’’ and ‘‘SSD’’ stand for two differ-ent moisture states of recycled coarse aggregates: oven-dried and saturated-surface-dried states, respectively. Lastly, the ending number ‘‘0’’, ‘‘30’’, or ‘‘100’’ says thezpercentage of recycled aggregates in the total amount of coarse aggregates.

2.2. Materials

Type I Portland cement is used that has the specific gravity of 3.15 and thefineness of 3312 cm2/g. Fly ash has the specific gravity of 2.14 and the fineness of3360 cm2/g, and several other properties of fly ash are shown in Table 2. The

Table 1Test variables of the eleven mixture cases.

Mixturecasea,b

Designcompressivestrength (f 0c)

Moisturestate ofcoarseaggregatec

Replacementratio of recycledaggregate (%)d

Use of flyash (30% ofcement)

N-SSD-0 24 MPa SSD 0 XNF-SSD-0 0 ON-SSD-30 30 XNF-SSD-30 30 ON-SSD-100 100 XNF-SSD-100 100 ON-OD-100 OD 100 X

H-SSD-0 40 MPa SSD 0 XH-SSD-30 30 XH-SSD-100 100 XHF-SSD-100 100 O

a N, H: design compressive strength (f 0c) equal to 24, 40 MPa.b F: presence of fly ash as a partial substitution (30%) of cement.c SSD, OD: moisture state of coarse aggregate, Saturated-Surface-Dried, Oven-

Dried.d 0, 30, 100: percentage of recycled coarse aggregate in the total amount of coarse

aggregate.

properties of coarse and fine aggregates are presented in Table 3. The recycledcoarse aggregates used have a lower density and a higher water absorption ratio(equal to approximately 6%) than the natural coarse aggregates. This is likely be-cause of residual mortar existing at the surfaces of recycled aggregates [23], asshown in Fig. 3. The recycled aggregates have a lower density and a higher waterabsorption ratio than those recommended in Korean Concrete Standard Specifica-tions [10], but they satisfies related requirements specified by RILEM [24] (seeTable 3). (Note that the manufacturing process used to convert demolished concreteto recycled aggregates did not deliberately remove attached mortar.)

The particle size distributions of the used aggregates are presented in Fig. 2. Thenatural coarse aggregates are crushed aggregates (Fig. 3) that have the maximumsize of 25 mm. The maximum size of recycled coarse aggregates is also 25 mm,and the particle size distribution satisfies the recommendations in Korean ConcreteStandard Specifications [10]. The maximum size of fine aggregates is 10 mm.

2.3. Mix proportions

Table 4 summarizes mix proportions for the eleven concrete cases. The water/cement ratio of the mix design was 43% and 32% in the cases of the design compres-sive strength equal to 24 and 40 MPa, respectively. The percentage of fine aggregatewas 42% in all eleven cases. For the cases using fly ash, 30% of the cement wassubstituted by fly ash. A PC-type super-plasticizer was used to be 0.3% and 1% ofthe amount of binder in the cases of the design compressive strength of 24 and40 MPa, respectively.

3. Test methods

In this study, various tests for recycled aggregate concrete wereperformed such as flow curve test, compressive and tensilestrength tests, and chloride ion penetration test. Each test methodis briefly described in the following.

3.1. Flow curve test

In order to investigate effects of the test variables on the rheo-logical properties of concrete, the flow curve test was conductedusing the ICAR Rheometer system (Fig. 4). Also, the slump test, fol-lowed by ASTM C413, was carried out at the same time. The resultsfrom the two tests are compared.

The ICAR Rheometer is a testing system [25] capable of measur-ing the yield stress and plastic viscosity defined by the Bingham’smodel [26]. A brief description of the test procedures is in the fol-lowing. Fresh concrete was added to a 286 mm-diameter containerup to a height of 300 mm. Then, a vane with four blades was lo-cated at the center of the concrete container (see Fig. 4). Both thediameter and the height of the vane are 127 mm. The vane was ro-tated at a constant speed of 0.5 rev/s for the breakdown time(about 20 s) in order to maintain the viscosity of the concrete. Afterthe breakdown time, the rheometer measured the rotation speedof the vane and the torque applied to the vane, as the rotationspeed was reduced from 0.5 to 0.05 rev/s over 30 s. A total of sevenmeasurements, each at every 5 s for a set of the rotation speed andthe torque, were recorded. Using the test results, the linear rela-tionship between the rotation speed and the torque of the vanemay be obtained by regression analysis. At last, the yield stressand plastic viscosity of the flesh concrete is estimated accordingto the Bingham’s model represented by:

s ¼ s0 þ l _c ð1Þ

Here s is shear stress (Pa), s0 is yield stress (Pa), l is plastic viscosity(Pa s), and _c is shear strain rate (1/s).

3.2. Compressive and splitting tensile strength tests

Cylindrical specimens for compressive and splitting tensilestrength tests were 100 mm in diameter and 200 mm in height,fabricated according to ASTM C943-10. The compressive strengthtests were conducted following ASTM C39 at the age of 7, 28,and 91 days. The splitting tensile strength tests followed ASTMC496 at the age of 28 and 91 days.

Table 2Properties of fly ash.

SiO2

(%)Water(%)

Loss onignition(%)

Density(g/cm3)

Blainefineness(cm2/g)

Flowvalueratio (%)

Activityindex(%)

48.8 0.1 3.5 2.14 3360 101 81

502 K. Kim et al. / Construction and Building Materials 48 (2013) 499–507

3.3. Chloride ion penetration test

The chloride ion penetration resistance is often used as animportant indicator to measure the long-term durability of con-crete. The tests followed the NT BUILD 492 method [27] to evaluateeffects of the test variables on the chloride diffusion coefficient ofconcrete. This test method is well known as an efficient way ofmeasuring the diffusion coefficient within a short time of about24 h.

Disk-shaped specimens that are 100 mm in diameter and50 mm in thickness were installed on a device shown in Fig. 5,and chloride ions were forced to penetrate into the concrete spec-imens using an electric field for 24 h. After this process, the con-crete specimens were split into half, and a silver nitrate solutionwas sprayed on the split surfaces. The change of color on thesprayed surfaces was observed, and the average chloride penetra-tion depth was estimated. The chloride diffusion coefficient wascalculated using the equations below:

D ¼ RTzFE� xd � a

ffiffiffiffiffixdp

tð2Þ

E ¼ U � 2L

ð3Þ

a ¼ 2

ffiffiffiffiffiffiffiffiRTzFE

r� erf�1 1� 2cd

c0

� �ð4Þ

Here D is the non-steady-state migration coefficient (m2/s), z is theabsolute value of ion valence (=1 for chloride), F is the Faraday con-stant (=9.648 � 104 J/V mol), R is the gas constant (=8.314 J/K mol),T is the average of initial and final temperatures in the anolyte solu-tion (K), xd is the average chloride penetration depth (m), and t isthe test duration (s). Also, U is the absolute value of the appliedvoltage (V), L is the thickness of specimen (m), erf�1 is the inverseof error function, cd is chloride concentration at which the colorchanges (N), and c0 is chloride concentration in the catholyte solu-tion (N).

Table 3Properties of aggregates.

Property of aggregate Quality standard for recycled coarseaggregate (KCI [9])

U

Ra

Absolute dry density (g/mm3) More than 2.5 (more than 2.0)a 2

Absorption (%) Less than 3.0 (less than 10)a 6

Abrasion (%) Less than 40 2Absolute volume (%) More than 55 50.08 mm Sieve passing (%) Less than 1.0 0Alkali aggregate reaction Harmless HAmount of clay mass (%) Less than 0.2 0Stability (%) Less than 12 4Contents of

impurity (%)Organicimpurity

Less than 1.0 (volume) L(

Inorganicimpurity

Less than 1.0 (weight) L(

a Recommendations by RILEM [24].

4. Test results

4.1. Rheological properties

The results obtained from the flow curve tests of normalstrength concrete (f 0c = 24 MPa) are presented in Figs. 6 and 7.Fig. 6 shows the flow curves of the three concrete mixtures notcontaining fly ash (N-SSD-0, N-SSD-30, and N-SSD-100), whileFig. 7 shows those of the three concrete mixtures using fly ash(NF-SSD-0, NF-SSD-30, and NF-SSD-100) as a partial substitution(i.e., 30%) of cement. In each figure, the only test variable is the ra-tio of recycled coarse aggregates. For each mixture case, a variationin the applied torque to rotate the vane at different speeds is illus-trated. In general, the lower torque value at a certain rotationspeed means the greater flowability.

As shown in Figs. 6 and 7, the use of recycled coarse aggregatesappeared to increase the flowability of concrete; that is, it gener-ally resulted in lower torque values. This trend was more evidentin the cases containing fly ash (Fig. 7). In the cases not using flyash (Fig. 6), however, the concrete with 30% recycled coarse aggre-gates (N-SSD-30) required higher torque values than the concretewith natural aggregates only (N-SSD-0), when the vane rotationspeed was faster than about 0.2 rev/s.

In the cases not including fly ash (Fig. 6), the torque values ran-ged from 3.5 to 7.1 N-m for the considered range of the vane rota-tion speed. In contrast, for the cases with fly ash (Fig. 7), the torquevalues ranged from 2.6 to 5.5 N-m, smaller than those without flyash. This indicates that the use of fly ash generally improved theflowability of flesh concrete.

The measured values of slump are presented in Table 5. Theslump of concrete increased when the higher ratio of recycledaggregates was used, and the rate of increase of slump was similarregardless of the presence of fly ash. For example, the concretewith 100% recycled coarse aggregates showed 65–70% higherslump values than the concrete with natural coarse aggregatesonly (compare N-SSD-100 vs. N-SSD-0, NF-SSD-100 vs. NF-SSD-0). With regard to the effect of fly ash, the cases using fly ash asa 30% substitution of cement showed 45–100% higher slump valuesthan the cases with no fly ash. Therefore, it was concluded thatboth recycled coarse aggregates and fly ash are helpful in improv-ing the flowability of fresh concrete, as found from the flow curvetests (Figs. 6 and 7).

There are a couple of possible reasons why the use of recycledaggregates improved the flowability of flesh concrete. First, thehigher volume of absorbed water in the pores of mortars attached

sed aggregates Test regulation

ecycled coarseggregate

Natural coarseaggregate

FineAggregate

.14 2.62 2.56 KS F 2503 (ISO 6783 &7033)a

.28 0.84 1.41 KS F 2503 (ISO 6783 &7033)a

1.1 14.6 – KS F 25087 59 58 KS F 2527.6 0.2 1.6 KS F 2511armless KS F 2545.15 0.08 0.4 KS F 2512.9 2.4 3.5 KS F 2507ess than 1.0volume)

– – KS F 2576

ess than 1.0weight)

– –

Fig. 3. Coarse aggregates used in the tests: (a) recycled, and (b) natural.

Table 4Mix proportion for each mixture case (kgf/m3).

Mixture case Water Cement Natural coarse aggregate Fine aggregate Recycled coarse aggregate Fly ash AEa

N-SSD-0 159 369 1060 691 – – 1.11NF-SSD-0 159 258 1060 691 – 75 1.11N-SSD-30 159 369 742 691 273 – 1.11NF-SSD-30 159 258 742 691 273 75 1.11N-SSD-100 159 369 – 691 911 1.11NF-SSD-100 159 258 – 691 911 75 1.11N-OD-100 159 369 – 691 911 – 1.11H-SSD-0 159 495 1034 614 – – 4.95H-SSD-30 159 495 724 614 266 – 4.95H-SSD-100 159 495 – 614 888 – 4.95HF-SSD-100 159 347 – 614 888 101 4.95

a AE = air-entraining admixture.

Fig. 4. ICAR Rheometer system [25].

Fig. 5. Test device for measurement of chloride ion penetration (NT Build 492 [27]).

K. Kim et al. / Construction and Building Materials 48 (2013) 499–507 503

to the recycled aggregates increased the total amount of water inthe concrete. Second, the relatively round shape and low densityof the recycled aggregates likely reduced frictional resistance, com-pared to the angular shape of the crushed natural aggregates (seeFig. 3). The two characteristics of the recycled aggregates possiblyaffected the flowability of concrete.

From the flow curve tests, the yield stress and plastic viscosityof each mixture are determined as shown in Table 5. When the ra-tio of recycled aggregates increased, the yield stress decreased

while the slump increased. Fig. 8 illustrates the relationship be-tween slump and yield stress for the six normal strength cases(f 0c = 24 MPa). A roughly negative logarithmic relationship wasfound between the slump and yield stress of the three cases with-out fly ash (N-SSD series), and between those with fly ash (NF-SSDseries); in each series, the ratio of recycled aggregates varied only.

Fig. 6. Flow curves of the concrete mixtures not containing fly ash.

Fig. 7. Flow curves of the concrete mixtures containing fly ash.

Table 5Results of slump and flow curve tests.

Mixture case N-SSD-0

N-SSD-30

N-SSD-100

NF-SSD-0

NF-SSD-30

NF-SSD-100

Slump (mm) 77 88 131 116 182 190Yield stress (Pa) 1180 833 619 1016 780 590Plastic viscosity

(Pa s)21.4 86.3 75.5 27.8 10.0 32.2

Fig. 8. Relationship between slump (x) and yield stress (y).

504 K. Kim et al. / Construction and Building Materials 48 (2013) 499–507

(Two equations proposed by Wallevik [16] and Murata and Kuka-wa [28] are also plotted in Fig. 8. The equations are not directlycomparable with the test results, because they were determinedfor the mixtures with natural aggregates only and with other vari-ables such as water/cement ratio and curing time. It is likely thatmany factors should be taken into account in quantifying the rela-tionship between slump and yield stress.) On the other hand, theuse of fly ash only slightly reduced the yield stress of recycledaggregate concrete; no significant change of yield stress occurred,while slump greatly increased (i.e., 45–100%) with fly ash (com-pare N-SSD-30 vs. NF-SSD-30, and N-SSD-100 vs. NF-SSD-100).This difference may be because the lower plastic viscosity, pre-sented in the following, caused by fly ash contributed to the in-crease of slump.

As for the plastic viscosity of recycled aggregate concrete, noconsistent tendency is observed in the effect of recycled aggregateson the plastic viscosity of flesh concrete (compare N-SSD-30 vs.N-SSD-100, NF-SSD-30 vs. NF-SSD-100). On the other hand, therecycled aggregate concretes containing fly ash present signifi-cantly lower plastic viscosity values than those with no fly ash.However, no apparent segregation was noticed during the tests.In summary, fly ash increased the slump of recycled aggregateconcrete, seemingly by reducing the plastic viscosity withoutsegregation.

4.2. Compressive strength

The results of compressive strength tests for all eleven mixturecases (Table 6) are plotted in Fig. 9 (cases without fly ash) andFig. 10 (with fly ash). In general, the compressive strength of con-crete decreased when the higher ratio of recycled aggregates wasused. For the normal strength cases (f 0c = 24 MPa) without fly ash(Fig. 9), the use of 30% and 100% recycled aggregates (i.e., N-SSD-30 and N-SSD-100) caused approximately 3% and 13% reductionsin the compressive strength respectively, compared to thespecimen with natural aggregates only (N-SSD-0) at 28 days ofcuring. Similarly, for the normal strength cases with fly ash(Fig. 10), NF-SSD-100 showed an approximately 19% decrease inthe strength at 28 days of curing, compared with NF-SSD-0. Thestrength reduction caused by recycled aggregates was more appar-ent in the high strength cases (f 0c = 40 MPa). The specimens with30% and 100% recycled aggregates (H-SSD-30 and H-SSD-100)had 8% and 34% lower compressive strengths respectively, thanH-SSD-0 at 28 days of curing. However, it is noted that all caseswith 30% recycled aggregates showed only slightly smaller com-pressive strengths.

The use of fly ash as a 30% substitution of cement generallycaused a reduction in the compressive strength of recycled aggre-gate concrete, but the reduction appears not critical on a long-termperspective. The specimens NF-SSD-0, NF-SSD-30, and NF-SSD-100(Fig. 10) showed 3%, 7%, and 11% lower compressive strengths thanN-SSD-0, N-SSD-30, and N-SSD-100 (Fig. 9) respectively, at 91 daysof curing. On the other hand, the strength difference was much lar-ger at 7 days of curing: 47%, 43%, and 30% lower strengths in thesame order, respectively. This reflects that the inclusion of fly ashtypically slowed down the hydration process of concrete, leadingto the slower development of early strength over time. However,the use of fly ash did not affect the strength of concrete designedfor f 0c = 40 MPa with 100% recycled aggregates (compare H-SSD-100 vs. HF-SSD-100).

All the specimens with less than 30% recycled aggregates, evenwith fly ash, satisfied their respective design compressivestrengths (24 or 40 MPa) at 28 days of curing. RILEM [24] suggeststhat the minimum ratio of natural aggregate should be at least 80%,if coarse aggregate is a blend of recycled and natural aggregates.However, the test results show that using recycled aggregate up

Table 6Results of compressive and splitting tensile strength tests.

Mixture case Compressive strength(MPa)

Splitting tensile strength(MPa)

Days of curing Days of curing

7 28 91 28 91

N-SSD-0 30 31 35 3.5 3.6N-SSD-30 30 30 30 2.9 3.2N-SSD-100 27 27 28 2.3 2.6NF-SSD-0 16 27 34 2.5 2.7NF-SSD-30 17 27 28 2.2 2.8NF-SSD-100 19 22 25 2.2 2.4N-OD-100 15 18 18 1.2 1.3H-SSD-0 40 50 55 4.2 4.3H-SSD-30 37 46 47 3.2 3.5H-SSD-100 27 33 34 2.6 2.9HF-SSD-100 28 29 34 2.8 3.0

Fig. 9. Compressive strengths of the cases not containing fly ash (curing time inparenthesis: 7, 28, or 91 days).

Fig. 10. Compressive strengths of the cases containing fly ash (curing time inparenthesis: 7, 28, or 91 days).

Fig. 11. Splitting tensile strengths of the cases not containing fly ash (curing time inparenthesis: 28 or 91 days).

Fig. 12. Splitting tensile strengths of the cases containing fly ash (curing time inparenthesis: 28 or 91 days).

K. Kim et al. / Construction and Building Materials 48 (2013) 499–507 505

to 30% was acceptable to achieve the design compressive strengthof concrete. In contrast, the specimens with 100% recycled coarseaggregates achieved lower strengths than the design values, exceptfor the normal strength case with no fly ash (N-SSD-100).

The effect of moisture state of coarse aggregates on the com-pressive strength was investigated for the cases designed forf 0c = 24 MPa with 100% recycled aggregates (see Fig. 9). The con-crete with oven-dried recycled aggregates (N-OD-100) had 44%

and 36% lower strengths than the concrete with saturated-sur-face-dried recycled aggregates (N-SSD-100) at 7 and 91 days ofcuring, respectively. A possible reason for the significant strengthdrop is that the mixing water immigrated into the oven-dried recy-cled aggregates with a high absorption capacity. The lack of mixingwater likely affected the hydration process of concrete in a nega-tive manner.

4.3. Splitting tensile strength

The results of splitting tensile strength tests for all eleven mix-ture cases (Table 6) are shown in Fig. 11 (cases without fly ash) andFig. 12 (with fly ash). The tensile strength of concrete generallydecreased when the ratio of recycled aggregates increased, as ob-served in the compressive strength. However, the degree of reduc-tion in the tensile strength was greater than in the compressivestrength. For example, in the normal strength cases (f 0c = 24 MPa)that did not include fly ash (Fig. 11), the use of 30% and 100% recy-cled aggregates (N-SSD-30 and N-SSD-100) caused approximately17% and 34% reductions in the tensile strength respectively,compared to the case with natural aggregates only (N-SSD-0) at28 days of curing. Furthermore, the high strength cases(f 0c = 40 MPa) with 30% and 100% recycled aggregates (H-SSD-30and H-SSD-100) had 24% and 38% lower tensile strengths respec-tively, than H-SSD-0 at 28 days of curing. The greater strengthreduction in tension than in compression is well illustrated inFig. 13; all data points for recycled aggregate concrete are plotted

Fig. 13. Relationship between compressive and tensile strengths (Note: a linearregression line is drawn for each series; fctm = mean tensile strength, andfck = specified characteristic compressive strength, defined in fib MC2010 [29]).

Fig. 14. Chloride diffusion coefficients.

506 K. Kim et al. / Construction and Building Materials 48 (2013) 499–507

below the curve specified in fib MC2010 [29], while those for nat-ural aggregate concrete are mostly above the curve. This trendmight be possibly because existing defects (e.g., micro-cracks) inthe recycled aggregates affected more on the tensile strength thanon the compressive strength [30].

The use of fly ash as a 30% substitution of cement generallycaused a reduction in the tensile strength of the normal strengthspecimens (f 0c = 24 MPa); the smaller reduction occurred in the casewith the higher ratio of recycled aggregates. For example, NF-SSD-0, NF-SSD-30, and NF-SSD-100 (Fig. 12) showed 29%, 24%, and 4%lower tensile strengths than N-SSD-0, N-SSD-30, and N-SSD-100(Fig. 11) respectively at 28 days of curing, and 25%, 13%, and 8%lower strengths at 91 days of curing. However, the high strengthspecimen (f 0c = 40 MPa) having fly ash (HF-SSD-100) showed ahigher tensile strength than the specimen with no fly ash (H-SSD-100).

As for the effect of moisture state of coarse aggregates, the nor-mal strength concrete with oven-dried recycled aggregates (N-OD-100) had 48% and 50% lower tensile strengths than the concretewith saturated-surface-dried recycled aggregates (N-SSD-100) at28 and 91 days of curing, respectively (see Fig. 11); the percentageswere slightly higher than those in the compressive strength.

4.4. Chlorine ion diffusion coefficient

Fig. 14 plots chloride diffusion coefficients of the seven nor-mal strength cases (f 0c = 24 MPa) in Table 1. In all the cases, the

chloride diffusion coefficient decreased (i.e., the chlorideresistance increased) as the curing age of concrete progressedfrom 28 to 91 days. In the three cases with no fly ash (N-SSD-0, N-SSD-30, and N-SSD-100), the higher ratio of recycled aggre-gates downgraded the resistance to chloride ion penetration atthe age of 91 days, but this effect was not found at the age of28 days.

As for the effect of fly ash, the three cases containing fly ash(NF-SSD series) exhibited much higher chloride resistance thanthose not containing fly ash (N-SSD series) at the age of 91 days;the average chloride diffusion coefficient of the N-SSD series wasapproximately 15.0 � 10�12 m/s2, while that of the NF-SSD serieswas approximately 5.1 � 10�12 m/s2. At the age of 28 days, how-ever, the average chloride diffusion coefficient of the N-SSD serieswas 22.5 � 10�12 m/s2, and that of the NF-SSD series was21.4 � 10�12 m/s2. This indicates that the chloride diffusion coeffi-cient of NF-SSD series reduced by about 76% from 28 to 91 daysof curing, while that of N-SSD series reduced by about 33% only.This was likely because fly ash was finer than cement, so that flyash not only improved the particle size distribution, but also filledmicro pores in the ITZs around aggregates. The beneficial role offly ash was not effective at the age of 28 days, possible becausethe presence of fly ash typically delayed the hydration process ofconcrete.

5. Conclusions

This study investigated the mechanical and durability proper-ties of recycled aggregate concrete, such as the rheological proper-ties of fresh concrete, compressive strength, tensile strength, andchloride diffusion coefficient. The main test variables includedthe ratio (0%, 30%, and 100%) of recycled aggregates in the totalamount of coarse aggregates, the presence of fly ash as a partialsubstitution (30%) of cement, the design compressive strength(24 or 40 MPa) of concrete, and the moisture state (SSD or OD) ofrecycled coarse aggregates. In particular, the combined effects ofrecycled aggregates and fly ash were explored, which should beconsidered during the mix design. The findings and conclusionsmay be summarized as follows:

� The higher ratio of recycled coarse aggregates generally resultedin the better flowability of flesh concrete, which was found fromthe results of both flow curve and slump tests. A roughly nega-tive logarithmic relationship was found between the slump andyield stress of the cases with different ratios of recycledaggregates.� The use of fly ash improved the flowability of concrete. The

slump of concrete greatly increased due to fly ash, while theyield stress only slightly decreased. The recycled aggregate con-cretes containing fly ash presented significantly lower plasticviscosity values than those not containing fly ash.� The strength test results showed that the higher ratio of recy-

cled aggregates generally caused the lower compressive andtensile strengths of concrete. However, the cases using 30%recycled aggregates showed only small reductions in the com-pressive strength; all the specimens with less than 30% recycledaggregates (even with fly ash) satisfied their respective designcompressive strengths. In contrast, the negative effect of recy-cled aggregates was greater in the tensile strength than in thecompressive strength.� The presence of fly ash caused a reduction in the compressive

strength of recycled aggregate concrete, but the reductionswere not critical on a long-term perspective (less than 12% atthe curing age of 91 days). As for the tensile strength of con-crete, higher strength reductions up to roughly 20% occurreddue to fly ash at the age of 91 days.

K. Kim et al. / Construction and Building Materials 48 (2013) 499–507 507

� The effect of recycled aggregates on the chloride penetrationresistance of concrete was not clearly identified. In contrast,the cases with fly ash exhibited much higher chloride resistancethan those without fly ash at the age of 91 days. This happenedeven in the cases with recycled aggregates.

Acknowledgments

This research was supported by the Ministry of KnowledgeEconomy, Korea, under the RIS (Regional Innovation System) Sup-port Program, supervised by the Korea Institute for Advancementof Technology. Also, support by Basic Science Research Programthrough the National Research Foundation of Korea (NRF) fundedby the Ministry of Education, Science and Technology (Grant No.2010-0022955) is gratefully acknowledged.

References

[1] Metz B, Davidson O, Coninck HCD, Loos M, Meyer LA, editors. IPCC specialreport on carbon dioxide capture and storage. Cambridge, UnitedKingdom: Cambridge University Press; 2005. p. 442.

[2] United States Environmental Protection Agency. <http://www.epa.gov/region1/solidwaste/cnd>.

[3] Korean Ministry of Construction and Transportation. In: Report for currentstatus of natural aggregate in Korea. Korean Ministry of Construction andTransportation, Sejong Special Governing City, Korea; 2001.

[4] Mehta PK. Concrete technology for sustainable development. Concr Int1999;21:47–53.

[5] Schokker AJ. The sustainable concrete guide – strategies andexamples. Farmington Hills, Michigan, USA: US Green Concrete Council; 2010.

[6] Task Group 3.8 within Commission 3. Guidelines for green concrete structures(fib Bulletin No. 67). Lausanne, Switzerland: International Federation forStructural Concrete; 2012.

[7] Korea Environment Corporation. <http://www.cwms.or.kr/>.[8] Northeast Waste Management Officials’ Association. Construction and

demolition waste management in the northeast in 2006, Boston,Massachusetts, USA; 2009.

[9] Korean Ministry of Environment. Report for statistics of recycling inKorea. Sejong Special Governing City, Korea: Korean Ministry ofEnvironment; 2005.

[10] Korean Concrete Institute (KCI). Korean Concrete StandardSpecifications. Seoul, Korea: Korean Concrete Institute; 2009.

[11] Kong D, Lei T, Zheng J, Ma C, Jiang J, Jiang J. Effect and mechanism of surface-coating pozzolanic materials around aggregate on properties and ITZmicrostructure of recycled aggregate concrete. Constr Build Mater2010;24:701–8.

[12] Kou SC, Poon CS. Properties of self-compacting concrete prepared with coarseand fine recycled concrete aggregates. Cem Concr Compos 2009;31:622–7.

[13] Poon CS, Kou SC. Properties of concrete prepared with PVA-impregnatedrecycled concrete aggregate. Cem Concr Compos 2010;32:649–54.

[14] Nassar PUD, Soroushian P. Strength and durability of recycled aggregateconcrete containing milled glass as partial replacement for cement. ConstrBuild Mater 2012;29:377–86.

[15] Siddique R. Performance characteristics of high-volume class F fly ashconcrete. Cem Concr Res 2004;34:487–93.

[16] Wallevik JE. Relationship between the Bingham parameters and slump. CemConcr Compos 2006;36:1214–21.

[17] Xiao J, Li J, Zhang C. Mechanical properties of recycled aggregate concreteunder uniaxial loading. Cem Concr Compos 2005;35:1187–94.

[18] Kwan WH, Ramli M, Kam KJ, Sulieman MZ. Influence of the amount of recycledcoarse aggregate in concrete design and durability properties. Constr BuildMater 2012;26:565–73.

[19] Sriravindrarajah R, Wang NDH, Ervin LJW. Mix design for pervious recycledaggregate concrete. Int J Concr Struct Mater 2012;6(4):239–46.

[20] McNeil K, Kang THK. Recycled concrete aggregates: a review. Int J Concr StructMater 2013;7(1):61–9.

[21] Berndt ML. Properties of sustainable concrete containing fly ash, slag andrecycled concrete aggregate. Constr Build Mater 2009;23:2606–13.

[22] Sim J, Park C. Compressive strength and resistance to chloride ion penetrationand carbonation of recycled aggregate concrete with varying amount of fly ashand fine recycled aggregate. Waste Manage 2011;31:2352–60.

[23] A.C.I. Committee 555. Removal and reuse of hardened concrete. FarmingtonHill, Michigan, USA: American Concrete Institute; 2001.

[24] RILEM TC 121-DRG. Specification for concrete with recycled aggregate. MaterStruct 1994;27:557–9.

[25] Germann Instruments Inc. ICAR Rheometer Manual. Evanston, Illinois, USA;2007.

[26] Robertson RE, Stiff Jr HA. An improved mathematical model for relating shearstress to shear rate in drilling fluids and cement slurries. SPE J 1976;16:31–6.

[27] Nordic Council of Ministers. Chloride migration coefficient from non-steady-state migration experiments (NT Build 492). Espoo, Finland: Nordtest; 1999.

[28] Murata J, Kukawa H. Viscosity equation for fresh concrete. ACI Mater J1992;89(3):230–7.

[29] Fédération Internationale du Béton. Fib Model Code 2010 (MC2010). Lausanne,Switzerland: International Federation for Structural Concrete; 2012.

[30] Casuccio M, Torrijos MC, Giaccio G, Zerbino R. Failure mechanism of recycledaggregate concrete. Constr Build Mater 2008;22:1500–6.